Skip to content Skip to navigation

City Nature

The City Nature project at Stanford blends humanistic and social science approaches with "hard" science–in this case, satellite imagery for land cover and spatial analysis–in a research inquiry motivated by an important environmental and societal problem. It is estimated that global urban population will double in the next 40 years, with most of that growth occurring in moderately sized cities. City planners face critical challenges in providing adequate access to parks and open space and in conserving critical local habitat and environmental quality generally. All are well understood to be important factors for quality of urban life. In order for researchers and scholars, government professionals, planners, builders, and nonprofit organizations to better develop heuristics or best practices for planning and constructing nature in cities, it is important to study current conditions in cities, and seek explanatory factors for the considerable variation in both quantity and quality of "city nature."

The City Nature project team has so far developed two novel datasets and two applications we expect will help this community of interest as well as scholars to develop research questions and answer them. The project also provided valuable research experience for a dozen Stanford undergraduates over the spring, summer, and fall of 2012. For the first application, "Naturehoods," we developed statistical measures of "greenness" and "pavedness" for 34 of the largest cities in the United States. We merged these data layers with several demographic data variables and developed interactive mapping and visualization tools that permit exploration of the data. In the second application, we assembled the complete text from comprehensive plans from 37 of the largest cities in the United States, and used the natural language processing method, Latent Dirichlet Allocation (LDA), to analyze the topic content of the corpus as a whole, the 37 individual plans, and sections within each plan. Preliminary results can be explored in an interactive tool.